OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ebxml-bp message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [ebxml-bp] CD 2.0.2 identifiers and filenames


Team,
As a followup to Dale's email, the Committee Draft will become v2.0.2. 
We will enter this version (as specified originally in the CD ballot 
approved) into the 15-day public review.  I'll also place these 
assumptions on the web site when I refactor the Table of Contents on our 
ebxmlbp public page.  For more details, see the meeting minutes (24 Jan 
2006). [1]

Stephen, once you have returned from UBL meeting, please review these 
and see if any questions are left unanswered. Thank you.

[1] Posted at: 
http://www.oasis-open.org/apps/org/workgroup/ebxml-bp/document.php?document_id=16438

>moberg 1/24/2006: Here are the preferences and conventions that the TC was working under
>as far as I can recall.
>
>1. The namespace value will not change until the specification version
>changes (or at least this is the current intent).
>2. The namespace will contain a part "2.0" that matches the
>specification version value.
>3. The TC will follow the artifact conventions for the filenames and
>associated URLs as far as they are understood by the TC editors.
>4. The numbering on files names will have a "major.minor.release" format
>5. The specificationVersion value will be "2.0"
>6. The namespace URL will resolve to the most current approved schema
>file (using redirection)
>7. The most current schema always will have the schemaLocation value for
>its namespace as the URL ending with the schema artifact filename whose
>final pattern is:
>major.minor.release.'xsd'
>-----Original Message-----
>From: Stephen Green [mailto:stephen_green@bristol-city.gov.uk] 
>Sent: Wednesday, January 18, 2006 7:05 AM
>To: ebxml-bp@lists.oasis-open.org
>Subject: [ebxml-bp] CD 2.0.2 identifiers and filenames
>
>ebBP TC,
>
>Greetings. Just to note that the change of version identifier
>from 2.0.1 to 2.0.2 will require changes in the UBP, UBL
>SBS (spec), etc. The namespace is, as I understand it, immune
>from this change (deliberately so) since the namespace just
>has 2.0 (http://docs.oasis-open.org/ebxmlbp/ebbp-2.0) 
>but how about the schema names? The existing convention
>seems to be to make the main schema "ebbp-2.0.2.xsd" and
>signal schemas "ebbp-signals-2.0.2.xsd". To me this seems preferable. 
>However, I'd just note (following the convention we are trying in 
>UBL 2) that if there is the plan to keep instances fully compatible 
>with future minor versions of the schemas, then this may be 
>helped if the filename is as immune to version minor changes 
>as the namespace; in this case calling the schema
>"ebbp-2.0.xsd" makes it unnecessary to specify which minor
>(?.?.x) version is being used as the schema for a particular instance.
>
>The latter may not be a good idea for ebBP since an instance is not 
>necessarily compatible with past minor version schemas (for
>example instances having the externalDocumentDefRef attribute
>will not be valid against schemas from before minor version 2.0.2)
>so it may be better for ebBP to be encourage instances to be 
>more precise when pointing to the ebBP schemas by including the
>full version identifier in the filename. A factor to consider, with UBL
>hindsight, is whether XSD derivation will be used to ensure such
>backwards compatibility of future minor version schemas. If not
>then that might, I think, favour again the fully qualified (2.0.2)
>version in the schema filename.
>
>If the schema files will be called <something>-2.0.2.xsd, I would hope
>that the final edit to the schemas would leave them as just 
>"ebbp-2.0.2.xsd", "ebbp-signals-2.0.2.xsd", etc. If not, might I plead 
>that their filenames be established as soon as possible to allow the
>editing of the examples and UBL UBP definitions to make them
>accurate for the final filenames. 
>
>[I note that with ebBP the backwards-compatible minor versions
>are presently identified after the second period (x in ?.?.x) - not
>so with UBL (x in ?.x).]
>
>Many thanks.
>All the best
>Steve
>



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]